VerdictPal
Join

Models / OpenAI

GPT-5.6 Terra

GPT-5.6 · Released 2026-07-09

OpenAI's generally available balanced GPT-5.6 model, positioned below Sol for efficient everyday work at $2.50 / $15 per 1M tokens.

Balanced GPT-5.6 tier for daily coding, research, and agent workflows when Sol's higher price is unnecessary.

#9 of 168 on AA Index · current snapshot
57 AA Index · 15 benchmark rows · 3 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite57
ContextUnlisted
Input / 1M$2.5
Output / 1M$15
Knowledge cutoffunpublished
Output speed108 t/s
TTFT (default API)94.25s
Statussolid

Family profile

Best published score in each covered benchmark family.

74/100 avg
Knowledge
47
Reasoning
93
Coding
54
Agentic
88
Multimodal
81
Long context
80

6 tested benchmark families

Editor's note
Terra moved from limited preview to general availability on July 9, 2026. Independent benchmark rows now cover the family, while OpenAI still has not published the context limit.
Benchmark placements

Where GPT-5.6 Terra places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
AA time to first tokenPerformance#3/ 186Artificial Analysis2026-09-01
CritPtReasoning#3/ 1230Artificial Analysis — CritPt2026-09-01
Terminal-BenchAgentic#5/ 18488Artificial Analysis2026-09-01
Terminal-Bench-ScienceAgentic#5/ 99Terminal-Bench-Science 0.1 announcement2026-08-27
MMMU-ProMultimodal#7/ 881Artificial Analysis — MMMU-Pro2026-09-01
Vending-Bench 2Agentic#7/ 10Andon Labs Vending-Bench 22026-08-04
AA-LCRLong context#9/ 17980Artificial Analysis2026-09-01
AA-OmniscienceKnowledge#12/ 1247Artificial Analysis — AA-Omniscience2026-09-01
Artificial Analysis Intelligence IndexReasoning#13/ 18757Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#16/ 10640Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#18/ 18043Artificial Analysis2026-09-01
GPQA DiamondReasoning#19/ 18593Artificial Analysis2026-09-01
SciCodeCoding#19/ 18054Artificial Analysis2026-09-01
AA output speedPerformance#36/ 187Artificial Analysis2026-09-01
IFBenchReasoning#48/ 12771Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where GPT-5.6 Terra shows up, it's highlighted.

KnowledgeGPT-5.6 Terra · 47
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningGPT-5.6 Terra · 93
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingGPT-5.6 Terra · 54
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGPT-5.6 Terra · 88
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGPT-5.6 Terra · 80
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for GPT-5.6 Terra in the last 120 days. Older rows live in the full table below.

15 recent
BenchmarkFamilyScoreSourceDays ago
CritPtReasoning30%Artificial Analysis — CritPt5
AA-OmniscienceKnowledge46.8%Artificial Analysis — AA-Omniscience5
MMMU-ProMultimodal80.7%Artificial Analysis — MMMU-Pro5
AA-LCRLong context79.67%Artificial Analysis5
AA output speedPerformance108 t/sArtificial Analysis5
AA time to first tokenPerformance94.25sArtificial Analysis5
Artificial Analysis Intelligence IndexReasoning56.6%Artificial Analysis5
GPQA DiamondReasoning92.5%Artificial Analysis5
Humanity's Last ExamReasoning42.9%Artificial Analysis5
IFBenchReasoning71.22%Artificial Analysis5
SciCodeCoding53.9%Artificial Analysis5
τ³-Bench BankingAgentic40.21%Artificial Analysis5
Terminal-BenchAgentic88.01%Artificial Analysis5
Terminal-Bench-ScienceAgentic8.6%Terminal-Bench-Science 0.1 announcement10
Vending-Bench 2Agentic7343.21Andon Labs Vending-Bench 233
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge47
Reasoning93
Coding54
Agentic88
Multimodal81
Long context80
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for GPT-5.6 Terra. Scores link to the original source; gaps mean no public row exists yet.

15 sourced rows

15 of 37 catalog benchmarks have a sourced row for GPT-5.6 Terra.

Knowledge

Reasoning

Coding

Agentic

Multimodal

Long context

Performance

Changelog
  1. Recorded general availability across ChatGPT, Codex, and the API and refreshed benchmark evidence.
  2. Updated public pricing: Terra is listed at $2.50 input / $15 output per 1M tokens.
  3. Previewed as the balanced GPT-5.6 tier.
Sources